The effect of d-cycloserine on negative symptoms (e.g., emotional withdrawal), quality of life and performance of cognitive tests sensitive to prefrontal cortex in 60 outpatients with schizophrenia who meet criteria for primary deficit syndrome and who have clinically significant negative symptoms (measured on the Scale for Assessment of Negative Symptoms (SANS) >40). Dysfunction of glutamatrergic neuronal systems has been implicated in the pathophysiology of schizophrenia based on the findings that non-competitive inhibitors of the NMDA receptor can reproduce in normals the positive symptoms, negative symptoms and cognitive deficits of schizophrenia. Furthermore, glutamatergic dysfunction may alter forebrain dopaminergic neuronal activity, a system central to the antipsychotic action of typical neuroleptics. We hypothesized that enhancing NMDA receptor function by systemic treatment with D-cycloserine, a partial agonist at the glycine modulatory site of the NMDA receptor, would reduce symptoms in schizohrenia. We have completed a double-blind dose finding study in a cohort of 10 patients with schizophrenia receiving typical neuroleptics and found significant reduction in negative symptoms and improvement in performance on a frontal cortical cognitive task at a D-cycloserine dose of 50 mg daily. We also have data from 25 patients who have completed an eight-week placebo-controlled parallel group trial of 50 mg D-cycloserine added to typical neuroleptics, which replicate the previous findings. We now propose a larger, six month placebo-controlled trial to characterize further the effects of D-cycloserine augmentation on negative symptoms, performance on neurocognitive tasks, and on markers for glutamatergic, dopaminergic and serotonergic function in serum and cerebrospinal fluid. This study will provide the power to determine if negative symptoms and cognitive function continue to improve over time, if these improvements meaningfully impact quality of life factors, if they correlate with markers of neuronal function, and if subpopulations can be identified according to response. We will randomly assign 60 schizophrenic outpatients with prominent, primary negative symptoms to D-cycloserine 50 mg or placebo for a six-month, fixed dose trial. The primary outcome measure is the total score on the Scale for Assessment of Negative Symptoms (SANS). We will administer a neuropsychological battery which emphasizes tests sensitive to prefrontal cortical function. Blood will be obtained at several time points and CSF will be obtained at baseline and week eight for assay of concentrations of d-cycloserine, glutamate, glycine, asparate, HVA and 5HIAA.